[Debian-med-packaging] Bug#653408: Bug#653408: [RFR] templates://biomaj/{templates}

[Olivier Sallou]

Le 12/29/11 7:08 PM, Justin B Rye a écrit :
> Christian PERRIER wrote:
>>  Template: biomaj/jdk_home
>>  Type: string
>> +_Description: JDK home directory:
>> + Please enter the directory where the Java Development Kit,
>> + which is used by BioMAJ, is installed.
> That's European-style relative clause punctuation.  You want:
>
>     Please specify the directory where the Java Development Kit
>     which is used by BioMAJ is installed.
>
> Or, since the place where it's installed is the place where it is,
> simplify to:
>
>     Please specify the path to the Java Development Kit for BioMAJ.
>
>> Give more details. I however wonder why this has to be
>> configured. Shouldn't the package depends on a JDK and therefore
>> "know" where it is installed?
> It does have a "Depends: openjdk-6-jdk|default-jdk|sun-java6-jdk", so
> yes, this sounds like a job for policy and/or $JAVA_HOME.
In fact, JDK path is required somewhere in configuration files, that's
why question is asked, even if jdk dependency is set.
>
>>  Template: biomaj/mysql
>>  Type: boolean
>> +_Description: Configure MySQL connection now?
>> + Please choose whether you want to configure connection
>> + parameters for the BioMAJ MySQL database.
>> + .
>> + Please note that the database has to be created and configured
>> + before it can be used.
>>
>> Use "BioMAJ" as this seems to be the "official" way to write the
>> software name?
> Upstream seem to think so.  I notice there's a bit more information in
> the revised version, but it's still rather unclear to me exactly what
> it is that needs to happen first or what this allows to happen second.
>
> Candidates for both include:
> a) the package mysql-server-x.y being fully installed
> b) a "biomaj_log" database and dbadmin being created
> c) the connection parameters being set up via debconf
> d) BioMAJ being configured to populate the database automatically
>
> I'm guessing that (a) is a prerequisite for (c) which gives you (d),
> but is doing (b) also a prerequisite or is it automated?

(a) is a prerequisite
For automation, (b) and (c) are required for (d).
If (b) or (c) is not satistifed, then database population has to be done
manually after package installation.

> Maybe:
>
>     Once the MySQL database for BioMAJ has been created and configured, it
>     can populate itself automatically instead of needing to be updated
>     manually.
>     .
>     Please specify whether the database connection should be configured now.
>   
>>  Template: biomaj/mysql_host
> [...]
>>  Template: biomaj/mysql_login
> [...]
>>  Template: biomaj/mysql_passwd
> These look good.
>
>>  Package: biomaj
> [...]
>> +Description: biological banks updater
> The top hits for "biological banks" are all talking about tissue
> sample repositories.  What you mean is
>
>    Description: biological data-bank updater
Correct
>
>> + BioMAJ downloads remote data banks, checks their status and applies transformation
>> + workflows, with consistent state, to provide ready-to-use data to biologists and
>>   bioinformaticians. For example, it can transform original fasta files to blast
>>   indexes. It is very flexible and post-processes can be extended very easily.
>>
>> No leading uppercase in synopsis.
>>
>> Some grammar corrections.
> Better, but: s/fasta/FASTA/, s/blast/BLAST/ (not to be confused with
> the Debian package "blast"), and I'm not sure about "post-processes" -
> maybe "its post-processing facilities"?
Post-processing sounds better indeed.
The software execute some programs, defined in a property file, after
the download of the data. Those programs are referred here as the
"post-processes".
>
>     BioMAJ downloads remote data banks, checks their status and applies
>     transformation workflows, with consistent state, to provide ready-to-use
>     data for biologists and bioinformaticians. For example, it can transform
>     original FASTA files into BLAST indexes. It is very flexible and its
>     post-processing facilities can be extended very easily.
>   
>>  Package: biomaj-properties
> [...]
>>  Description: bank properties file examples
> Even worse - "bank properties" means land owned by financial
> institutions...
>
>    Description: biological data-bank updater - example properties files
>
>>   This package contains a set of bank properties that can be used as
>> + examples for BioMAJ. It contains many existing banks descriptions
>> + with or without workflows to transform data.  Some files may require
>> + additional programs for post-processes such as blast.
> A couple of extra tweaks (for instance, they're existing data banks,
> not existing descriptions):
>
>     This package contains a set of data bank properties files that can be
>     used as examples for BioMAJ. It contains descriptions for many existing
>     data banks with or without workflows to transform data. Some files may
>     require additional programs (such as BLAST) for post-processing.
>
>
>
> _______________________________________________
> Debian-med-packaging mailing list
> Debian-med-packaging at lists.alioth.debian.org
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-packaging

-- 
Olivier Sallou
IRISA / University of Rennes 1
Campus de Beaulieu, 35000 RENNES - FRANCE
Tel: 02.99.84.71.95

gpg key id: 4096R/326D8438  (pgp.mit.edu)
Key fingerprint = 5FB4 6F83 D3B9 5204 6335  D26D 78DC 68DB 326D 8438


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/debian-med-packaging/attachments/20111230/8389e5c5/attachment-0001.html>